The advancement in technology and availability of Computer Mediated Communication (CMC) technology and the Internet made meaningful interaction possible in both on and offline environments, which is referred to as blended learning. Blended learning has recently begun to gain popularity in ELT, and appropriate models of blended learning need to be developed for EFL classrooms. However, there have not been many attempts to investigate its application and development of appropriate teaching and learning models of blended learning in the field of ELT, especially in relation to English language skills and subskills. Therefore, the purposes of this study are to investigate the students’ perspectives and effectiveness of blended learning in English pronunciation skills, and to explore gender differences in blended learning. The subjects of this study consist of 173 Korean university students, and quantitative data including questionnaire and pre-experiment were collected. The major findings of the study are as follows. First, students’ perspectives in blended learning, its components, and tools were very positive. Second, blended learning was found to be effective in the development of segmental and suprasegmental pronunciation skills. Third, there were few significant gender differences found regarding social aspects of online interaction. Implications and suggestions based on the findings and discussion are provided.